I call this model SDXL_NSFW_Clean:
The reason for the clean is that I spent approximately 127 hours sampling images and output images in order to get very clean results. The trained LoRA that I used to make the Base Model had a loss of 0.1508 as the LoRA was created with a learning rate of .0004000, 20 steps after it passed through 10,000 steps at a noise level setting of 0. The Base Model does not need any LoRA to produce nice result as it functions very nicely by itself, but LoRAs are what make generating images fun.
Description
I call this model SDXL_NSFW_Clean:
The reason for the clean is that I spent approximately 127 hours sampling images and output images in order to get very clean results. The trained LoRA that I used to make the Base Model had a loss of 0.1508 as the LoRA was created with a learning rate of .0004000, 20 steps after it passed through 10,000 steps at a noise level setting of 0. The Base Model does not need any LoRA to produce nice result as it functions very nicely by itself, but LoRAs are what make generating images fun.
FAQ
Comments (2)
So I gather this one works better with lora, but is otherwise just SDXL, correct?
Correct
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.